Speaker and Phoneme-Aware Speech Bandwidth Extension with Residual Dual-Path Network.
Nana HouChenglin XuVan Tung PhamJoey Tianyi ZhouEng Siong ChngHaizhou LiPublished in: INTERSPEECH (2020)
Keyphrases
- speech recognition
- automatic speech recognition
- speech synthesis
- automatic speech recognition systems
- speaker dependent
- speech signal
- speaker identification
- speaker recognition
- network bandwidth
- audio visual
- prosodic features
- vocal tract
- speech sounds
- bandwidth allocation
- network resources
- multiple paths
- communication links
- phoneme recognition
- hidden markov models
- speaker verification
- network traffic
- speech recognizer
- multicast tree
- gaussian mixture model
- link failure
- communication networks
- network model
- path selection
- visual speech
- network congestion
- optical networks
- speech recognition systems
- wireless sensor networks
- language model
- atm networks
- bandwidth usage
- speaker independent
- asynchronous transfer mode
- pattern recognition
- shortest path
- network structure
- bandwidth requirements
- transmission rate
- allocation scheme
- traffic load